System Health APIs: which one to use
MDSS exposes three health endpoints on the API gateway. They answer different questions, so pick the one that matches what you want to know.
Endpoint | Question it answers | Credentials | Available |
|---|---|---|---|
| Is the API gateway process up? | None from 4.6.0. Before 4.6.0, the | All versions |
| Can MDSS serve requests right now? | None | From 4.7.0 |
| How is MDSS performing, and why? |
| All versions |
Which one should I use?
A load balancer or monitor that needs up or down.
GET /api/readyzon MDSS 4.7.0 or later. On 4.6.x useGET /api/health; it stays200during a database, cache or broker outage, so it detects only a dead gateway. On 4.5.x and earlier/api/healthneeds an API key; a Read Only Administrator key is enough.Kubernetes. Liveness probe on
/api/health, readiness probe on/api/readyz.A script that waits for MDSS to start. Poll
/api/healthuntil the process answers, then/api/readyzuntil MDSS is ready to log in and scan.Slow scans, a pause or throttling.
GET /api/healthzwith an administrator key, or the System tab in the UI. It explains the state and recommends an action.
What each endpoint sees
|
|
| |
|---|---|---|---|
API gateway process | yes | yes | yes |
Cache, database and message broker connections | no | yes | no (database capacity only) |
Discovery, Scanning and Remediations reporting | no | yes | yes |
Storage guardrail pause and throttling | no | yes, as | yes, with reasons and limits |
MetaDefender Core instances | no | yes, | yes, per instance |
Host resources, throughput, recommendations | no | no | yes |
Never point a health check at /. It is a static page served by the web server and answers 200 even while the API is down.